近红外光谱-BP神经网络-PLS法用于橄榄油掺杂分析

橄榄油兼有食用和保健的作用,价值与价格远远高于其他食用油,所以橄榄油中以劣充好的现象十分普遍。可采用近红外光谱法测定初榨橄榄油中掺杂芝麻油、大豆油和葵花籽油的光谱数据,运用改进的BP算法——Levenberg-Marquardt方法,建立PCA-BP人工神经网络方法对其进行定性判别。同时采用偏最小二乘法(PLS)建立了初榨橄榄油中芝麻油、大豆油、葵花籽油含量的近红外光谱定标模型,用交互验证法进行验证。结果表明,BP人工神经网络有很好的定性鉴别能力,PLS建立的芝麻油、大豆油、葵花籽油定标模型的相关系数分别为98.77,99.37,99.44,交叉验证的均方根误差分别为1.3,1.1,1.04。该方法无损、快速、简便,为橄榄油掺杂的检测提供了一种新的途径。     

基于拉曼光谱和最小二乘支持向量机的橄榄油掺伪检测方法研究

为实现橄榄油中掺伪油类型的识别和掺伪量预测,对掺入葵花籽油、大豆油、玉米油的橄榄油共117个样品进行拉曼光谱检测,并用基于多重迭代优化的最小二乘支持向量机模型对掺入油的类型进行识别,综合识别率为97%。同时分别采用最小二乘支持向量机、人工神经网络模型、偏最小二乘回归建立橄榄油中葵花籽油、大豆油、玉米油含量的拉曼光谱定标模型,结果显示最小二乘支持向量机具有最优的预测效果,其预测均方根误差(RMSEP)在0.007 4～0.014 2之间。拉曼光谱结合最小二乘支持向量机可为橄榄油掺伪检测提供一种精确、快速、简便、无损的方法。     

拉曼光谱结合模式识别方法用于大豆原油掺伪的快速判别

大豆原油是我国的战略储备物资,然而目前储油市场上频繁出现大豆原油掺混的现象严重影响了食用油储备安全。基于此,通过大豆原油与部分植物精炼油拉曼谱图的特征差异,并结合主成分分析-支持向量机(PCA-SVM)模式识别建立了大豆原油是否掺伪的快速判别方法。以28个大豆原油、46个精炼油、110个掺伪油的拉曼谱图为模型样本;选择位于780～1 800 cm-1波段的谱图,预处理方法同时采用Y轴强度校正、基线校正和谱图归一化法;在此基础上应用PCA法提取特征变量,即以贡献率最高前7个主成分为变量进行SVM分析。SVM校正模型的建立是以随机选取的20个大豆原油和75个掺伪油样组成校正集,以8个大豆原油和35个掺伪油样组成验证集,分别运用并比较四种核函数算法建立的大豆原油SVM分类模型,并采用网格搜索法(grid-search)优化模型的参数,以四种模型的分类性能作为评判标准。结果表明：应用线性核函数算法构建的SVM分类模型可以很好地完成掺伪大豆原油的判别,校正集识别准确率达到100%,预测结果的误判率为0,判别下限为2.5%。结果表明应用拉曼光谱结合化学计量学能够用于大豆原油掺伪的快速鉴别。拉曼光谱简便、快速、无损、几乎没有试剂消耗,适合现场检测,从而为大豆原油的掺伪分析提供了一种新的备选方法。     

近红外光谱技术在食用油掺伪检测中的研究进展

食用油是日常饮食的必需品，可以为人体提供热能和脂肪酸，是促进脂溶性维生素吸收的重要有机物。随着人们生活水平的提高，高档食用油已走进大众百姓的餐桌，并深受欢迎和喜爱。由于高档食用油市场售价高，一些不法厂商为牟取暴利，在高档食用油中掺入廉价食用油进行出售，导致食用油掺伪事件时有发生，已引起政府和民众的广泛关注。为保障消费者的合法利益和维护正常的食用油市场秩序，快速有效地检测食用油掺伪已刻不容缓。近红外光谱技术以其简便、快速、无损、无需样品预处理的特点，被广泛应用于食用油掺伪分析。概述了近红外光谱技术的基本原理，综述了近十年来近红外光谱技术在橄榄油、山茶油、芝麻油、核桃油等食用油的掺伪检测研究进展，包括采用不同的试验装置与试验方法、数据处理方法包括预处理、特征波长选择及建模方法，对二元、三元及多元食用油掺伪进行检测研究，从试验方法及数据处理等角度提高食用油掺伪检测的精度与适用范围，以期建立较为有效的食用油掺伪定量检测与定性鉴别模型。总结了食用油掺伪近红外光谱检测目前存在的问题，包括食用油掺伪检测机理不明晰，制备的掺伪食用油样本难以满足实际的复杂掺伪形式，采用取样方式的掺伪检测仅能实现现场部分...     

基于近-中红外相关谱融合判定掺伪芝麻油

为充分提取复杂掺伪食用油的特征信息,提出并建立一种掺伪芝麻油的判别方法.采集40个纯芝麻油和40个掺入不同浓度玉米油的芝麻油的常规一维近红外透射光谱和中红外衰减全反射光谱.对两样本采用二维相关谱技术进行相关计算,得到每一样品的同步和异步二维近红外相关谱和中红外相关谱,并进行预处理,得到其对应的同步-异步二维近红外相关谱和中红外相关谱.采用多维主成分分析法提取其特征,并将其得分矩阵进行融合.基于融合的得分矩阵,以及单一近红外、中红外相关谱得分矩阵分别建立纯芝麻油和掺伪芝麻油偏最小二乘判别分析模型,三个模型对预测集样品的判别正确率分别为100%、96.2%和96.2%.研究结果表明,所提出的方法可提取更多的特征信息,提供更好的分析结果.     

基于激光近红外的稻米油掺伪定性-定量分析

该文主要研究激光近红外光谱分析技术结合化学计量学方法对稻米油掺伪进行定性-定量分析。分别将大豆油、玉米油、菜籽油、餐饮废弃油掺入稻米油中,按照不同质量比配置189个掺伪油样,利用激光近红外光谱仪采集光谱;对采集的稻米油掺伪图谱数据进行多元散射校正(MSC)、正交信号校正 (OSC)、标准正态变量变换和去趋势技术联用算法(SNV_DT)三种不同预处理并与原始数据进行比较。采用连续投影算法(SPA)对经过预处理的光谱数据进行特征波长提取,应用支持向量机分类(SVC)方法建立稻米油掺伪样品的定性分类校正模型,选择网格搜索算法对模型参数组合(C,g)进行寻优,确定最优参数组合。另采用后向间隔偏最小二乘法(BiPLS)和SPA对预处理后的光谱数据进行特征波长提取,分别应用偏最小二乘法(PLS)和支持向量机回归(SVR)建立掺伪油含量的定量校正模型,并选用网格搜索算法对SVR模型参数组合(C,g)进行寻优,建立最优参数模型。研究表明,建立的SVC模型预测集和校正集的准确率分别达到了95%和100%;对比SVR和PLS方法建立的数学模型对稻米油中掺杂油脂的含量的预测,两种方法均能够实现含量预测,SVR模型的预测能力更好,相关系数R高于0.99,均方根误差(MSE)低于5.55×10-4,预测精度高。结果表明,采用激光近红外光谱分析技术可以实现稻米油掺伪的定性-定量分析,同时为其他油脂的掺伪分析提供了方法。     

基于拉曼光谱-模式识别方法对奶粉进行真伪鉴别和掺伪分析

奶粉的真伪和掺伪近年来受到广泛的关注,研究一种操作便捷,能准确、快速、全面鉴定奶粉品牌并实现奶粉掺假鉴别的新方法对于奶粉的质量控制具有重要的意义。为实现奶粉的真伪鉴别,采集三种品牌奶粉贝因美、飞鹤和雀巢的拉曼光谱,并利用拉曼谱图特征峰结合最近邻算法(nearest neighbor,NN)的模型对三种品牌奶粉进行识别,在10次交叉验证的基础上,平均识别率为99.56%。为实现奶粉的掺伪分析,将飞鹤奶粉与雀巢奶粉按不同质量比(0∶1,1∶3,1∶1,3∶1,1∶0)混合成五种掺伪奶粉,提取掺伪奶粉中的脂肪,采集脂肪样本的拉曼光谱,分别使用拉曼谱图特征峰结最近邻算法的模型和核主成分分析(kernel principal components analysis,KPCA)结合最近邻算法的模型对五种脂肪样本进行识别,10次交叉验证下的平均识别率分别为93.33%和98.89%,平均运算时间分别为0.085和0.104 s。实验证明：特征峰结合NN的算法可以快速实现对奶粉真伪的判别,但此算法不能很好的区分掺伪奶粉;拉曼光谱-KPCA-NN模型可以为奶粉的掺伪检测提供一种简便、准确、快速的方法。     

三维荧光光谱结合Zernike图像矩快速鉴别掺伪芝麻油

为了实现对掺伪芝麻油的快速鉴别,应用FS920荧光光谱仪测定样品的三维荧光光谱数据。将三维荧光光谱图视为灰度图,在没有任何预处理的前提下,直接应用Zernike图像矩提取三维光谱灰度图的特征信息,然后采用类平均法对特征信息进行聚类分析,从定性角度实现掺伪芝麻油的鉴别,并解析其组成成分。最后应用广义回归神经网络(GRNN)对掺伪样本的成分进行定量分析。聚类分析能够以很高的辨识率来识别掺伪芝麻油,并能够正确解析其组成成分。定量模型预测了2组掺伪样本中各成分的相对体积,其平均相对误差分别为2.23%,8.00%,9.70%和9.70%。分析结果表明,Zernike矩能够有效提取光谱的特征信息,光谱数据的Zernike矩特征结合聚类分析以及GRNN模型能够获得良好的定性和定量分析结果,为掺伪芝麻油的鉴别提供了一种新的方法。     

基于高效分离不皂化物的傅里叶红外光谱结合化学计量学鉴定芝麻油

芝麻油是日常生活中常用食用油之一,掺假芝麻油会导致严重的健康问题。研究芝麻油鉴定方法是非常重要的。皂化植物油提取不皂化物是食用油鉴定的经济方法之一,现有植物油皂化方法需要较长时间,较高的温度,且不皂化物提取过程非常繁琐。采用超声技术替代常规回流加热法,提高了皂化效率,皂化时间缩短至10分钟,在此基础上采用专用固相萃取（SPE）小柱快速分离不皂化物。基于分离富集得到的植物油不皂化物红外光谱,结合化学计量学方法进行芝麻油鉴定。利用偏最小二乘判别分析（PLS-DA）和正交偏最小二乘判别分析（OPLS-DA）构建芝麻油鉴定模型。分析结果表明：所构建的芝麻油鉴定模型,OPLS-DA模型优于PLS-DA模型;OPLS-DA模型对芝麻油检验集样本预测准确率高。基于植物油不皂化物红外光谱结合化学计量学方法可以准确鉴定芝麻油。     

小波矩结合三维荧光光谱简单快速鉴别掺伪芝麻油

芝麻油营养丰富,因市场价格较高,掺假现象频出,严重损害了消费者利益和市场的健康发展。因此,研发一种简单快速准确鉴别掺伪芝麻油的方法,对保障消费者权益和市场健康具有重要意义。为此,提出了一种小波矩结合三维荧光光谱掺伪芝麻油鉴别方法。该方法简单快速,计算样本的任一有效特征进行谱系聚类,即可准确鉴别掺伪芝麻油。以43个样本(芝麻油16个,掺伪菜籽油、掺伪大豆油及掺伪玉米油各9个)为研究对象,用FS920荧光光谱仪获得样本的三维荧光光谱。用db2小波将光谱进行多尺度分解(MRSD),用MRSD的一阶离散逼近系数构造小波矩。用前两阶小波矩值W_(0,0),W_(1,0),W_(1,1),W_(0,1),W_(2,0),W_(2,1),W_(2,2),W_(1,2),W_(0,2)分别作为特征对样本进行谱系聚类,观察分析聚类结果。结合邓恩分类指数(DVI)进一步分析,研究同阶小波矩分类效果及规律。进而研究各阶小波矩的分类效果及规律。最终确定了用于鉴别掺伪芝麻油的最佳小波矩值。结果表明:MRSD一阶逼近重构光谱可以在保留原光谱的有效特征基础上,大量去除噪声,减少光谱数据量72.4%,增强模型的抗噪稳定性和实时性。利用小波矩前两阶矩值W_(2,1),W_(2,2),W_(1,2),W_(0,2)其一作为分类特征进行谱系聚类,即可鉴别掺伪芝麻油。同阶小波矩(W_(p,q))随p值减小q值增大呈现规律性,确定了同阶小波矩的有效矩值及最佳有效矩值。小波矩随着阶数的增加DVI先增后减,最后趋于稳定,确定了各阶小波矩中可用于鉴别掺伪芝麻油的目标矩值W_(0,q≥2)及最佳目标矩值W_(0,6)。小波矩的有效及目标矩值是针对样本分类的有效特征,计算样本的任一有效特征进行谱系聚类,即可实现掺伪芝麻油的鉴别。该研究思路及结论为矩值法应用到三维荧光光谱提供参考。该方法简单快速,可实现在线测量,为质监部门及生产企业提供油品检测和鉴定手段。     

Analysis of Argan Oil Adulteration Using Infrared Spectroscopy

The determination of argan oil adulteration by other vegetable oils is a real analytical challenge. The authentication of argan oil needs fast and simple analytical techniques for quality control and testing. This study focuses on the detection and quantification of argan oil adulteration with different edible oils, using midinfrared spectroscopy with chemometrics. Chemometric treatment of MIR spectra has been assessed for the classification and quantification of argan oil adulteration with sunflower or soybean oils. The potential of MID spectroscopy combined with partial least squares regression (PLS) as a rapid analytical technique for the quantitative determination of adulterants in argan oil has been demonstrated. A PLS model has been established to predict the concentration of soybean and sunflower oil as adulterants in the calibration range between 0% and 30% (w/w) in argan oil with good prediction performances in the external validation.     

Adulteration of sesame oil with corn oil detected by use of two-dimensional infrared correlation spectroscopy and multivariate calibration

An innovative methodology was developed to detect adulteration of sesame oil with corn oil based on two-dimensional mid-infrared correlation spectroscopy with multivariate calibration. Forty pure sesame oils and 40 adulterated sesame oils with corn oil were prepared and the infrared absorption spectra were measured at room temperature, respectively. The synchronous two-dimensional mid-infrared correlation spectra were calculated to develop multivariate calibration models for adulteration of sesame oil with corn oil. The results showed the higher classification accuracy of 96.3% for the prediction set using two-dimensional mid-infrared correlation spectra and N-way partial least square discriminant analysis, versus 88.9% using traditional one-dimensional mid-infrared spectra and partial least squares discriminant analysis. Also, the multivariate calibration models were developed for quantitative analysis of sesame oil adulteration with corn oil. The root mean square error of prediction was 0.98% v/v using two-dimensional mid-infrared correlation spectra and N-PLS, and 1.15% v/v using traditional one-dimensional mid-infrared spectra and PLS. The results of our analyses indicated that the proposed method could provide better predictive results than traditional one-dimensional mid-infrared spectra and multivariate calibration.     

Visible Raman spectroscopy for the discrimination of olive oils from different vegetable oils and the detection of adulteration

We have investigated the potential of Raman spectroscopy with excitation in the visible spectral range (VIS Raman) as a tool for the classification of different vegetable oils and the quantification of adulteration of virgin olive oil as an example. For the classification, principal component analysis (PCA) was applied, where 96% of the spectral variation was characterized by the first two components. A significant similarity between sunflower oil and extra‐virgin olive oil was found using this approach. Therefore, sunflower oil is a potential candidate for adulteration in most commercially available olive oils. Beside the classification of the different vegetable oils, we have successfully applied Raman spectroscopy in combination with partial least‐squares (PLS) regression analysis for very fast monitoring of adulteration of extra‐virgin olive oil with sunflower oil. Different mixtures of extra‐virgin olive oil with three different sunflower oil types were prepared between 5 and 100% (v/v) in 5% increments of sunflower oil. While in the present context the adulteration usually refers to the addition of reasonable amounts of the adulterant (given the similarity with the basic product), we show that the technique proposed can also be used for trace analysis of the adulterant. Without using techniques like surface‐enhanced Raman scattering (SERS), a quantitative detection limit down to 500 ppm (0.05%) could be achieved, a limit irrelevant for adulteration in commercial terms but significant for trace analysis. The qualitative detection limit even was at considerably lower concentration values. Based on PCA, a clear discrimination between pure extra‐virgin olive oil and olive oil adulterated with sunflower oil was achieved. The adulterant content was successfully determined using PLS regression with a high correlation coefficient and small root mean‐square error for both prediction and validation. Copyright © 2009 John Wiley & Sons, Ltd.     

基于最小二乘支持向量机的橄榄油掺杂拉曼快速鉴别方法

提出了一种基于最小二乘支持向量机(LS-SVM)的橄榄油掺杂拉曼快速鉴别方法。首先,收集若干己知类别的橄榄油样作为训练样本,获取其拉曼谱图,并对其谱图进行预处理和波段选择,进而构建LSSVM分类器;对于未知类别的油样,获取其拉曼谱图,并进行相应的预处理和波段选择,由LSSVM分类器获得鉴别结果。实验以7种已知的特级初榨橄榄油为基础,分别掺入4种其它植物油(大豆油、菜籽油、玉米油、葵花籽油),获得112个掺杂油样。将全部样本随机分成训练集和测试集,对测试集样本的预测实验结果表明,本文方法能有效鉴别橄榄油掺杂,且掺杂量最低检测限为5%。与其它分类方法相比,LSSVM分类法具有最佳的分类性能。该方法快速、简便,为橄榄油掺杂鉴别提供了一种全新的方法。     

同步-异步二维中红外相关谱检测掺假芝麻油

将同步-异步二维中红外相关谱和多维偏最小二乘判别法相结合定性分析掺假芝麻油。分别配置40个纯芝麻油和含有玉米油不同体积分数(3%~60%)的掺假芝麻油样品40个。室温下,分别采集所有样品的常规一维中红外光谱(650~4 000cm-1)。在研究纯芝麻油和掺假芝麻油的一维中红外光谱的基础上,以芝麻油中掺假的玉米油浓度为外扰,进行相关计算,得到同步和异步二维中红外相关谱矩阵,并对其进行标准化。分别提取标准化的同步和异步二维中红外相关谱主对角线上部分和下部分元素进行融合,得到同步-异步二维中红外相关谱矩阵。在此基础上,分别基于同步-异步二维中红外相关谱矩阵、同步二维中红外相关谱矩阵和异步二维中红外相关谱矩阵建立了三个定性分析掺假芝麻油的多维偏最小二乘判别模型对预测集未知样品进行预测,其识别正确率分别为100%,96.2%和96.2%。结果表明:相对于同步和异步二维中红外相关谱,同步-异步二维中红外相关谱不仅包含了完整的掺假油特征信息,而且剔除了冗余信息,因此能取得更好的判别结果。     

衰减全反射红外光谱法快速分类和识别多种食用油的方法

提出了一种衰减全反射红外光谱法快速分类和识别多种食用油的方法——KL-BP模型。此模型利用KL算法对原始光谱数据分类特征进行提取并对原始数据降维,降维后的数据作为神经网络的输入建立分析模型。实验共收集了九种食用油包括芝麻油、玉米油、油菜籽油、调和油、葵花油、花生油、橄榄油、大豆油、茶籽油,共84个样品,并测定了其衰减全反射红外光谱。为了对比所提方法性能,分别建立PCA直接分类、KL直接分类、PLS-DA、PCA-BP和KL-BP模型的分类结果进行对比。研究结果表明,对所研究的9种食用油,PCA直接分类、KL直接分类、PLS-DA、PCA-BP和KL-BP方法的识别率分别为59.1%,68.2%,77.3%,77.3%和90.9%。在数据降维中,KL算法通过分别提取使类间距离和类内距离比值最大方向的特征向量提取和包含在类内离散度矩阵中的分类信息,能够比PCA方法提取了更多的分类信息;引入BP神经网络能有效地提高分类能力和分类准确率;KL-BP综合了KL对分类信息提取优势以及BP神经网络自学习、自适应、非线性的优点,在分类和识别成分相近的9种食用油中表现出了最优秀的能力。     

The use of infrared spectroscopy in combination with chemometrics for quality control and authentication of edible fats and oils: A review

Edible fats and oils provide a significant contribution in our diet and daily life, as cooking or frying oil, or as components used in food, pharmaceutical, and cosmetics products. Fats and oils are characterized by specific values, including acid value, saponification value, iodine value, and peroxide value, as well as the oxidation products which occur during storage due to oxidative and hydrolytic deterioration. Currently, due to the high price of edible fats and oils, some unethical producers adulterate high-value edible oils like olive oil with low-priced oils like palm and corn oils; therefore the authentication analysis of edible fats and oils must be assured by introducing reliable and fast methods like infrared spectroscopy. Fourier transform infrared (FTIR) spectroscopy is an ideal technique for monitoring the quality control of fats and oils due to its property as a “fingerprint spectra technique,” which allows analysts to differentiate among fats and oils. FTIR spectra signals of fats and oils are very complex. Fortunately, a statistical technique called chemometrics can be used to handle the complex FTIR spectral data. Chemometrics in combination with FTIR spectroscopy has been widely used in many aspects of monitoring quality control of edible fats and oils including their authenticity.     

Quantitative detection of adulterated olive oil by Raman spectroscopy and chemometrics

Commercially available extra virgin olive oils are often adulterated with some other cheaper edible oils with similar chemical compositions. A set of extra virgin olive oil samples adulterated with soybean oil, corn oil and sunflower seed oil were characterized by Raman spectra in the region 1000–1800 cm⁻¹. Based on the intensity of the Raman spectra with vibrational bands normalized by the band at 1441 cm⁻¹ (CH₂), external standard method (ESM) was employed for the quantitative analysis, which was compared with the results achieved by support vector machine (SVM) methods. By plotting the adulterant content of extra virgin olive oil versus its corresponding band intensity in the Raman spectrum at 1265 cm⁻¹, the calibration curve was obtained. Coefficient of determination (R²) of each curve was 0.9956, 0.9915 and 0.9905 for extra virgin olive oil samples adulterated with soybean oil, corn oil and sunflower seed oil, respectively. The mean absolute relative errors were calculated as 7.41, 7.78 and 9.45%, respectively, with ESM, while they were 5.10, 6.96 and 4.55, in the SVM model, respectively. The prediction accuracy shows that the ESM based on Raman spectroscopy is a promising technique for the authentication of extra virgin olive oil. The method also has the advantages of simplicity, time savings and non‐requirement of sample preprocessing; especially, a portable Raman system is suitable for on‐site testing and quality control in field applications. Copyright © 2011 John Wiley & Sons, Ltd.